The endoplasmic reticulum (ER) is the major site of protein biosynthesis in eukaryotes. Polypeptides entering the ER may occasionally adopt aberrant conformations, resulting in aggregation-prone, misfolded proteins. The accumulation of misfolded proteins represents a form of ER stress, which has been implicated in the pathogenesis of many human diseases. To preserve ER homeostasis, eukaryotes have evolved a conserved quality control pathway termed retro-translocation or dislocation, which efficiently eliminates unwanted proteins from the ER by exporting them into the cytosol. Polypeptides undergoing retro-translocation are disposed of by the cytosolic proteasome. The retro-translocation pathway is hijacked by certain viruses to destroy folded cellular proteins required for immune response, allowing the virus to evade host immune surveillance. The molecular mechanism of retro-translocation is largely unknown. For example, it is not well understood how cells can distinguish misfolded polypeptides from those that are in the folding process. How misfolded substrates are selectively targeted to the translocation site at the ER membrane, and subsequently transferred across the membrane are completely unknown. The identity of the protein-conducting channel for retro-translocation is still under debate. In addition, how viruses can exploit this cellular pathway during their invasion into the host cell is unclear. We have previously identified a cytosolic enzyme called p97, which provides the major driving force to move substrates into the cytosol during retro-translocation. Two co-factors of p97, Ufd1 and Npl4, are also required. The ATPase complex interacts in its ATP bound state with substrates emerging from the ER membrane, and the two ATPase domains appear to alternate in ATP hydrolysis to release polypeptides from the ER membrane once they are modified by poly-ubiquitination. Interestingly, we found that the ATPase complex contains several ubiquitin binding domains that specifically recognize ubiquitin chains. This partially explains why the ATPase complex preferentially acts on poly-ubiquitinated substrates. The interaction between the ubiquitin chains and p97 may trigger ATP hydrolysis by the ATPase, allowing it to pull substrates out of the ER membrane. To understand how p97 functions at the ER membrane, we used an affinity purification approach to identify two novel ER membrane proteins, Derlin-1 and VIMP, which associate with p97. VIMP functions as a receptor to recruit p97 to the ER membrane. The conserved multi-spanning membrane protein Derlin-1 plays a central role in retro-translocation, perhaps as a component of the protein-conducting channel. It receives substrates from the ER lumen, and also associates on the cytosolic side of the ER membrane with both the ubiquitination machinery and the "pulling" ATPase p97. Thus, it provides a link between substrate recognition in the ER lumen and polypeptide dislocation in the cytosol. We also demonstrated that efficient elimination of misfolded ER proteins also involves a p97-associated deubiquitinating enzyme, ataxin-3. Mutations in ataxin-3 have been linked to type-3 spinocerebellar ataxia, a member of the poly-glutamine induced neurodegenerative diesease family, but the physiological function of ataxin-3 is unclear. We show that overexpression of an ataxin-3 mutant defective in deubiquitination inhibits the degradation of misfolded ER proteins and triggers ER stress. Misfolded polypeptides stabilized by mutant ataxin-3 are accumulated in part as poly-ubiquitinated form, suggesting an involvement of its deubiquitinating activity in ERAD regulation. We demonstrate that ataxin-3 transiently associates with the ER membrane via p97 and the recently identified Derlin-VIMP complex, and its release from the membrane appears to be governed by both the p97 ATPase cycle and its own deubiquitinating activity. We present evidence that ataxin-3 may promote p97-associated deubiquitination to facilitate the transfer of polypeptides from p97 to the proteasome. In the past year, we dissected the role of intramembrane charged residues in ER quality control of T- cell receptor. We found that a TCR mutant lacking the intramembrane charged residues has a tendency to form homo-oligomer via interchain disulfide bond that involves a specific pair of cysteine residues. Covalent oligomerization of TCR appears to stabilize it at the ER membrane. The presence of a single lysine residue at specific positions within the TCR TM domain abolishes its oligomerization and causes its rapid degradation. Conversely, when TCR oligomerization is induced by a bivalent compound, the degradation of TCR is inhibited. Together, these results suggest that the intramembrane charged residues in TCR do not function as a signal for substrate recognition in ERAD. Instead, their primary role is to reduce TCR oligomerization to maintain it in a retrotranslocation competent state. Our results also suggest that the ERAD machinery is inefficient when coping with oligomerized substrates, indicating a requirement for chaperone-mediated protein disassembly in the ER lumen prior to retrotranslocation. We also studied the mechanism by which the human cytomegalovirus (HCMV) protein US2 hijacks the ER-associated degradation (ERAD) machinery to dispose of MHC class I heavy chain (HC) at the endoplasmic reticulum (ER). We established an in vitro permeabilized cell assay that recapitulates the retrotranslocation of MHC HC in US2-expressing cells. Using this assay, we demonstrate that the dislocation process requires ATP and ubiquitin, as expected. The retrotranslocation also involves the p97 ATPase. However, the mechanism by which p97 dislocates MHC class I HC in US2 cells is distinct from that in US11 cells: the dislocation reaction in US2 cells is independent of the p97 cofactor Ufd1-Npl4. Our results suggest that different retrotranslocation mechanisms can employ distinct p97 ATPase complexes to dislocate substrates.